We observed slight discrepancies when running the tests on certain
machines.

Access OMF object by pos, length and chunk data - expert use only.
Background for this development is a request by Andrew. He likes the idea
of keeping AFDB outside Mongo in big data blobs. However, for each uniprot
AC it's OK to store a couple of numbers => the numbers that are stored in
the indexing mechanism of FSStructure server (pos, length, chunk). This can
be quickly changed in case of an update without massive data shuffling.
Accessing an OMF object from a cold data blob then requires to open the
memory mapped data and ask the OS to jump to a certain location in that
data blob. This avoids loading the whole indexing data if only a single
structure needs to be loaded.

Sidechains and BackboneLists now use bond length and angle parameters that are
extracted from a non-redundant set of high resolution X-ray structures.
IC data from CHARMM36 was used before. The new parameters slightly improve
sidechain modelling accuracy.

Modelling unit tests used a custom compound library. This compound
library can't be loaded with latest OST anymore. It's safe to assume
that the compound library which is required by ProMod3 does the job.

changes in OpenStructure CompoundLib required rebuild of existing dummy
compoundlib

The following issues were observed:
- warning about non-writable directory which was set as IPYTHONDIR env var
  Not very surprising, it's a directory that lives in container space...
- stale kernels
  I didn't investigate that one properly. It had something to do with bending
  around all Jupyter/IPython relevant directories to custom set directories
  in the image. Reason for doing that is potential Python modules that reside
  in user home of the host that get imported in the container. I removed the
  bending around with the full knowledge that the described problem may occur.
- version updates
  Some changes related to different Python versions between old base image
  (Ubuntu 20.04) and the new one (Ubuntu 22.04).

nglview requires ipywidgets < 8

adapt to changes in ost release-2.4.0 branch in conop processors

The Motif finder algorithm was designed to have a database of small motifs
and search them in one target structure. Needs some changes for the use case
of lots of larger structures in which we want to find 1 Motif.

'ling' instead of 'filling' also catches 'sampling' which is used in
CloseLargeDeletions desaster recovery.

Avoids triggering errors with more strict parameter checks introduced
in OpenMM 7.4